Comparative anticonvulsant potency and pharmacokinetics of (+)- and (−)-enantiomers of stiripentol

29

Epilepsy Research, 12 (1992) 29-36 0920-1211/92/$05,00@ 1992 Elsevier Science Publishers B.V. All rights reserved EPIRES 00479

Comparative anticonvulsant potency and pharmacokinetics of (+)- and (-)-enantiomers of stiripentol

Danny

Shena, Rem5

Levya, Jill

Savitcha, Alan Lepage”

Boddyb, Francis

and

“Department ofPharmaceutics, School of Pharmacy, University of Washington, Seattle, WA 98195 (USA), bPharmacogenetics Research Unit, Department of Pharmacological Sciences, The Medical School, University of Newcastle Upon Tyne, NE2 4HH (UK) and Yen&e de Recherche, Laboratoires Biocode*, 602UQCompiPgne (France) (Received 20 September 1991; revision received 1.5February 1992; accepted 28 February 1992) Key words: Stiripentol enantiomers; Pentylenetetrazol;

Stereoselectivity; Anticonvulsant potency; Pharmacokinetics

The anticonvulsant potency and pharmacokinetics of the enantiomers of stiripentol were compared using the intravenous pentylenetetrazol infusion seizure model in the rat. Enantioselectivity was observed with respect to both the anticonvulsant activity and elimination kinetics of this compound. (+)-Stiripentol was 2.4 times more potent than its antipode against pentylenetetrazol-induced clonic seizure (brain EC,, of 15.2,@ml versus 36.1 &ml). The (+)-enantiomer was eliminated more rapidly than the (-)-enantiomer, as reflected in a higher plasma clearance (1.64 l/h/kg versus 0.557 &/kg) and a shorter half-life (2.83 h versus 6.50 h). Parallel studies with the racemate of stiripentol indicated that the anticonvulsant potency of the racemate was between the potency of the two enantiomers, suggesting that the combined activity reflects the additive action of (+)- and (-)-stiripentol. However, a marked metabolic interaction between enantiomers was evident after racemate administration. These results point to the need for information on the differential pharmacokinetics of stiripentol enantiomers following racemic drug administration.

Stiripentol (l-[3,4-methylenedioxyphenyl]-4,4dimethyl-1-penten-3-01) is a new antiepileptic drug currently undergoing clinical trial. It was selected from a series of aromatic allylic alcohols which demonstrated anticonvulsant and hypnotic activities in rodent screens’. The pharmacology of stiripentol was recently reviewed by Loiseau and Correspondence to: Dr. Danny D. Shen, Department of Pharmaceutics, BG-20, University of Washington, Seattle, WA 98195, USA.

Duche14. A notable chemical feature of stiripentol is the presence of a chiral center at C-3. All previous pharmacologic testing has been performed with the racemic form of the compound. The pharmacodynamic characteristics of the individual stereoisomers have not been investigated. This communication presents the. results of an acute-dose-response study aimed at comparing the disposition kinetics and anticonvulsant potency of the two enantiomers of stiripentol by using the timed intravenous pentylenetetr~ol (PTZ) infusion seizure model in the rat. To relate the present data to those of an earlier efficacy study with

30 racemic stiripentol in the same animal model”, a comparison between the racemate and the individual enantiomers is included. Methods

Materiais Stiripentol enantiomers were synthesized from the corresponding acetylenic alcohols obtained by stereoselective reduction of the acetylenic ketone. Fig. 1 shows the synthetic scheme. The intermediate, 3,4-methylenedioxyphenylethynyl trimethylsilane (see Fig. 1, structure 1) was obtained either from 3,4_methylenedioxy benzaldehyde or from 3,4-methylenedioxy acetophenone. The trimethylsilane intermediate was then acylated with trimethylacetyl chloride under the Friedel-Crafts conditions as described by Babin et al.‘, yielding an ethynyl ketone (Fig. 1, structure 2). Asymmet-

1 tEUCOCl/AICI&!H~C1~.

-30°C

5 (+) 6 C-1 Fig. 1. A scheme for the stereoselective (-)-stiripentol.

synthesis of (+)- and

ric reduction of the ethynyl ketone was accomplished by a procedure described by Tanno and Terashima*’ using the chiral hydride originally prepared by partially decomposing lithium aluminum hydride with (-)- or (+)-N-methylephedrine and N-ethylaniline, to yield the (-)- or (+)-propargylic alcohols (Fig. 1, structure 3 or 4), which were then converted into their respective allylic alcohols without racemization. Optical purity of the stiripentol enantiomers was determined by reacting the synthetic products with naphthyl isocyanate to form the naphthyl carbamate derivative of (+)- or (-)-stiripentol, followed by resolution of the diastereomeric derivatives by chiral phase high-performance liquid chromatography (HPLC) using a Pirkle Type 1-A reversible column. The optical purity of the (+)-isomer used in this study (D 2232, lots 3 and 4) was >97% (enantiomeric excess). On the other hand, some batches of the (-)-isomer were contaminated with a variable, but small, quantity of the (+)-isomer (D 2233, lots 3,4, and 5 with respective optical purityof lOO%, 85%, and97%). Animals Male Sprague-Dawley rats reared in a specific pathogen free environment were obtained from Charles River, Inc. (Portage, MI). Their body weights ranged from 250 to 300 g. The rats were acclimatized in our animal holding facility on a 12h light/dark cycle for at least 48 h before being entered into the study. Food and water were available ad libitum throughout the study.

The first phase of the study was aimed at characterizing the absorption and elimination kinetics of stiripentol after a single intraperitoneal administration of the drug in the form of the enantiomers or the racemate. This information was needed for the design of the subsequent pharmacodynamic studies. A silastic catheter was surgically implanted into the right, external, jugular vein of the rat for serial blood sampling. The (+)- or (-)-enantiomer of stiripentol was dissolved in corn oil and administered by intraperitoneal injection to separate groups of rats (n = 3) at a dose of 300 mg/kg. A third group

31 of rats (n = 3) received the same dose of racemic stiripentol. Blood was withdrawn immediately before drug administration and at 5, 15, 30, 45, 60, and 90 min and at 2,4,6,8, and 24 h after drug administration. The blood samples were anticoagulated with sodium EDTA and centrifuged, and the plasma was stored at -70°C pending analysis. Pharmacodynamic

experiments

The ability of stiripentol to elevate the minimal (clonic) seizure threshold of PTZ was assessed by the timed intravenous infusion technique originally developed by Orloff et al.17 in the mouse and recently modified for use in the rat by Pollack and Shentg. There were two reasons for emplo~ng this seizure model. First, the early pharmacologic screens18, as well as a recent study in our laboratorY2? indicated that stiripentoi is highly effective against PTZ-induced seizures in rodents. Second, the timed intravenous PTZ infusion test has the advantage of providing graded measures of anticonvulsant effect in individu~ animals, which allows determination of plasma and brain concentration-effect relationship in a relatively small group. An efficient animal seizure model was needed for the present pha~a~dynamic study because of the limited supply of stiripentol enantiomers for in vivo testing. A siiastic catheter was surgically implanted into the right external jugular vein of the rats. The animals were allowed 24 h to recover from surgery before being tested for their baseline or pre-drug convulsive response to PTZ. Pentyl~netetr~ol (16 mg/ml in saline) was infused through the indwelling jugular vein catheter at a rate of 8 mg/min (0.5 mYmin) until the onset of a clonic convulsion. The threshold dose (mg/kg) of PTZ required to elicit the seizure end point was determined from the duration of PTZ infusion (i.e., TD,). On day 2, the rats were given a single intra~~toneal injection of one of the stiripentol enantiomers or of the racemate dissolved in corn oil. For each form of stiripentol, three or more dosage levels (between 300 and 900 mglkg) were tested. Based on the data from the preceding pharmacokinetic study, the time point of 1 h was chosen for the acute-dose-response study. This was the time when the plasma drug concentrations from a given

dose of the stiripentol enantiomers or the racemate were similar, which allowed the same dosage range to be used for all three forms of stiripentol. At 1 h after drug administration, a blood sample was drawn through the venous catheter. This was followed by a second PTZ infusion test to determine the post-stiripentol threshold dose of PTZ (i.e., TD,). To avoid diurnal variation in seizure response, the treatment protocol was arranged such that the PTZ tests were performed at the same time every day. The animals were decapitated immediately after seizure testing. The entire brain was removed, rinsed thoroughly with normal saline, blotted lightly to remove adhered blood, and stored at -7OY. Drug assays The plasma

concentration of stiripentol was measured by a reversed-phase HPLC method described by Lin and Levyt3. The interday coefftcient of variation of the procedure was consistently less than 6%. Frozen rat brain was sectioned sagitally into halves and weighed after thawing. One of the brain halves was homogenized in 4 times its volume of methanol. After centrifugation of the homogenates, aliquots (25-50 ~1) of the supematants were taken for stiripentol assay by another recently described HPLC procedure developed in our laboratoryro. The interday coefficient of variation of the brain assay was about 3%. Since these chromatographic procedures do not resolve the enantiomers of stiripentol, total concentration of the two enantiomers was measured in samples collected after racemic drug administration. Data analysis

Pharmacokinetic parameters describing the absorption, distribution, and clearance characteristics of the stiripentol enantiomers and the racemate were calculated based on plasma concentration-time data from the single dose disposition studies4. The absorption rate following intraperitoneal drug administration was indicated by the time to reach peak plasma concentration (T,). Time to peak and the maximum serum drug concentration (C,) were estimated by interpolation of the observed plasma concentration-time data using the

32

Lagrange formula7. The elimination half-life (Ti,& was calculated by least-squares regression analysis of the terminal log-linear portion of the plasma drug concentration-time curve (i.e., the last 3 time points), assuming that in all instances absorption rate was much more rapid than the elimination rate of the drug. The apparent plasma clearance (CL/F) was computed by dose/AUC, where AUC or area under the plasma concentration-time curve was estimated by the Lagrange approximation procedure described by Yeh and Kwax?. The apparent volume of distribution (V/F) was, in turn, calculated from clearance and half-life estimates (i.e., CL/F.1.44.T,,J. The anticonvulsant effect of stiripentol (E), as assessed by the increase in seizure threshold of PTZ, was expressed in terms of the logarithm of the threshold dose ratio or log,, (TDJTD,). The relationship between log,, dose ratio and the plasma or brain concentration at 1 h after administration of either of the stiripentol enantiomers or the racemate was analyzed by the sigmoidal effect-concentration model as described by Holford and Sheiner’: E=

J%ax*cY EC,,” + Cy

(1)

where E,, = maximum log,, dose ratio, C = plasma or brain drug concentration, EC,, = concentration at which 50% of maximal effect is achieved, and y = sigmoidicity . Our previous experience with the intravenous PTZ infusion seizure model in the rat indicated that the dose ratio cannot exceed 3.0, i.e., a maximum threefold elevation in the PTZ threshold dose. Continued infusion of PTZ beyond this ceiling inevitably led to either a sudden fatal clonictonic seizure complex or toxicity from PTZ (pulmonary hemorrhage). Therefore, E,, was set at a value of log,, 3 or 0.477 and EC, was, in effect, the observed drug concentration at a PTZ dose ratio of 1.73 or anti-log,, (0.477/2). Estimates of the pha~acodynamic parameters were obtained by non-linear least-squares regression fit of the concentration-effect data to Equation 1.

0

2

4

6

8

10

1: 14 Time (hr)

16

18

20

22

24

Fig. 2. Time course of piasma drug concentratjon following a single intraperitoneal dose (300 mg/kg) of (+)-stiripentol (O), (-)-stiripentol (0), and racemic stiripentol (x) in the rat.

Results

The mean plasma concentration-time profiles after intraperitoneal administration of the stiripent01 enantiomers, individually or as their racemate, are shown in Fig. 2. A comparison of the derived pharmacokinetic parameter estimates is presented in Table I. Peak plasma concentration was reached between 1.5 and 2.0 h after intraperitoneal administration of the two enantiomeric forms of stiripentel. There were no statistically signi~~ant differences in mean Tp. (+)-Stiripentol was eliminated much more rapidly than its antipode, as indicated by an apparent plasma clearance that was nearly three times higher, and by a shorter elimination half-life (P < 0.05). There was no statistically sig-

TABLE I A comparison of the ~har~coki~tic parameter stats ripentol when admin~tered in the form of ~~vid~f mers or as a racemate Parameter”

(+)-STP

(-)-STP

T, (h) C,,, @s/ml) CL/F (I/h/kg) V/F (l/kg) Tm (h)

1.50 + 39.0 + 1.64 f 6.68 + 2.83 t

1.75 * 27.9 It 0.557 rt 5.80 It 6.50 f

0.30b 18.0 0.37 0.10 0.59

of stierza&o-

(+)-s-f-P 0.10 2.00 + 0.40 5.7 40.0 + 10.6 0.236 0.276? 0.113 3.92 5.62 & 2.02 1.91 12.3 f 8.7

a See text for a de~nition of the ph~ma~~metic ’ Mean t- standard deviation (n = 3).

symbols.

33 nificant difference in the apparent volume of distribution (V/F) between the enantiomers. It should be noted that the observed difference in CL/F could be related to enantioselectivity in bioavailability rather than systemic clearance of stiripentol. However, the lack of a notable difference in the V/F estimates of the two enantiomers suggests that bioavailability is not a likely confounding factor. A comparison of pharmacokinetic parameter estimates, obtained after administration of the racemate, with those of the individual enantiomers suggested the possibility of an interaction between enantiomers. It was expected that after administration of the racemate, the total plasma concentration level of stiripentol would fall between the levels achieved after administration of the individual enantiomers, but in fact it was higher. This deviation is reflected in a lower apparent CL/F for racemic drug than for the individual enantiomers (P < 0.05). In addition, the decline in total stiripentol concentration after racemate administration was much slower than the decline of either of the enantiomers. Acute-dose-response study Concentrations of stiripentol in the plasma taken at 1 h following doses of 300-900 mg/kg of either the enantiomers or the racemate varied over a large range, from about 1 to 100 pg/ml. Even though a statistically significant correlation (2 = 0.41 to 0.85, P < 0.05) could be demonstrated be-

tween plasma concentration and dose for the three forms of stiripentol, the interanimal variability in plasma stiripentol concentration at a given dose of the drug was so large that the relationship of anticonvulsant effect to dose was ill-defined. Fig. 3 shows a plot of brain concentration versus plasma concentration of stiripentol at 1 h. The brain concentration of stiripentol was generally lower than that in plasma, ranging from less than 1 &ml to 70 ,@ml. Moreover, the curvilinear nature of the relationship between brain and plasma concentrations indicated that the partitioning of stiripentol across the blood-brain barrier was highly concentration-dependent; the brain-to-plasma partitioning ratio increased from about 0.25 to 0.70 as the plasma drug concentration was increased from 1 to 100 ,@ml. In addition, the partitioning ratios appeared to be the same regardless of whether the (+)- or (-)-enantiomer, or a racemic mixture of the two, was administered. The anticonvulsant response, expressed as log,, PTZ dose ratio, was plotted against either plasma or brain drug concentration. Fig. 4 shows a comparison of the brain concentration-response relationships obtained from the administration of either the (+)- or (-)-enantiomer of stiripentol. The inset of the figure shows a similar plot against

l

3.0 f

2.5

H

2.0

S

0

n

. 0

1.5

.

0

u ” Lo

. I’ I’ ,‘.I

$3 .

1.0

so

SO

Plasma Cont. (ue/mL)

Fig. 3. The relationship between brain and plasma concentrations of stiripentol at 1 h following the administration of 300-900 mgkg of (+)-stiripentol (0). (-)-stiripentol (0) and racemic stiripentol (x) in the rat.

liD

J!!?7k

I 0

40

0

*

/

0

*

P

*

o(

I. 4

20

40

60 30 Brain Cont. (u@J)

loo

120

Fig. 4. Brain concentration-effect relationship for (+)-stiripent01 (0) and (-)-stiripentol (0) in the timed intravenous PTZ infusion seizure model. The continuous tines represent the regression fit of the data to the sigmoidal concentration-response model equation. The inset shows a plot of effect against total brain drug concentration for racemic stiripentol (x). ‘Ihe dotted line represents the fitted response curves for the racemate data and is compared with the solid lines for the individual enantiomers.

34 TABLE II A comparison of the pharmacodynamic parameter estimates of stiripentol when adminbtered in the form of individual enantiomers or as a racemate (+)-UP

(-)-STP

(+)-S-P

EC, Cuplml) Y

43.0 f 3.2a 5.45 + 3.40

65.2 + 4.0 5.50 + 1.50

54.1 + 5.2 4.50 z!z1.60

Brain EC, b&s) Y

15.2 + 0.7 2.77 f 0.3

36.1 + 2.1 3.20 + 0.56

20.0 f 1.5 3.77 + 1.58

Parameter Plasma

a Mean + standard deviation of the nonlinear least-squares regression estimates of the parameters.

total plasma stiripentol concentration for the raceThe (+)-stiripentol-resmate administration. ponse curve lies parallel to and to the left of the (-)-stiripentol curve, indicating that the (+)-isomer is the more potent form of stiripentol. The pharmacodynamic parameter estimates are presented in Table II. Based on a comparison of their brain EC,, the (+)-enantiomer was 2.38 times more potent against PT.Z-induced clonic seizures in the rat. The response curve for the racemate was situated midway between the two enantiomer curves (see Fig. 4, inset). This suggests that the two stiripentol enantiomers act in an additive fashion when they are present simultaneously after administration of stiripentol racemate. A similar comparison was observed with the plasma-concentration-effect relationships (see Table II). However, the potency ratio of (+)-stiripentol compared with (-)-stiripentol based on plasma EC, was more modest at 1.52. The apparently smaller difference in the anticonvulsant potency between the enantiomers is due to the nonlinearity in the partitioning of stiripentol between brain and plasma; i.e., for a given difference in plasma concentrations, a greater difference is observed in their corresponding brain concentrations. Discussion Stereoselective metabolism has been demonstrated with a number of new and existing chiral antiepileptic agents, such as mephenytoin, ethotoin”, mephobarbital”, and vigabatrit?. Surpris-

ingly, there is much less information as to whether stereoisomerism also affects the anticonvulsant potency of the aforementioned antiepileptic compounds. For instance, while the literature is replete with examples of marked differences in the hypnotic and anesthetic potency between enantiomorphs of many barbiturates, to the best of our knowledge stereoselectivity in anticonvulsant properties has been investigated only with secobarbital in mice?. The most notable example that drew attention to the importance of stereoselectivity in pharmacodynamic action was the discovery that of the two enantiomers of vigabatrin only the S( +) form inhibits synaptosomal GABA-transaminase and exhibits anticonvulsant activitys*15. Most recently, Leander and his associates reported a marked degree of enantioselectivity in the dose-response characteristics of several new benzylamide anticonvulsants9”‘. Therefore, the present work represents one of the few coordinated attempts to describe the stereoselectivity of both the disposition and the pharmacologic potency of a new antiepileptic compound. Our data indicate that stiripentol enantiomers differ significantly in their anticonvulsant potency along with a remarkable difference in their plasma clearance characteristics. Since stiripentol is eliminated largely via metabolism23, the enantiomeric difference in plasma clearance can be attributed to a marked stereoselectivity in stiripentol metabolism in the rat. The distribution of stiripentol between brain and plasma was noted to be highly concentration-dependent, although the brain concentration versus plasma concentration curves for the enantiomers could be superimposed. Because of the nonlinear relationship, the corresponding enantiomeric concentration ratio in brain will differ slightly from that in plasma. The cause of the nonlinearity in brain distribution was not investigated. The increase in brain-to-plasma concentration ratios as the circulating level of stiripentol is elevated can be explained by a saturation in plasma drug protein binding. In view of the fact that racemic stiripentol is used in current clinical trials, it is useful to relate our present findings to the situation when a racemic mixture of the compound might be administered. The fact that the concentration-response curve

for racemic stiripentol fell between those of the enantiomers suggests that the anticon~ls~t activity is additive in nature when the two stiripentol enantiomers are present simultaneously. In other words, once the enantiomeric composition of stiripent01 in brain or plasma is known, the resulting activity is reasonably predictable. Consequently, the contribution of each enantiomer to the overall effect is dictated by the enantiomer’s individual pharmacokinetic characteristics. If we proceed on the a~umption that there is little or no pha~a~okineti~ interaction between the enantiomers (i.e., the disposition kinetics of each enantiomer are the same regardless of whether it is administered in pure form or as a racemate), the pharmacodynamics of racemic stiripentol in the rat can readily be extrapolated, based on the data from the present study. The difference in plasma half-life, observed after individual ad~nistration of the stiripentol enantiome~, suggests that the ~ncentration ratio of (-)-stiripentol to (+)-stiripentol could vary substantially over time after administration of the racemic compound. A rapid decline in the plasma concentration of the more potent (+)-enantiomer would lead to a predominant presence of the less active (-)-enantiomer at late times. In addition, a greater accumulation of (-)-stiripentol relative to (+)-stiripentol is expected during repetitive adm~istration of racemic stiripentol. Such changes in the relative abundance of the two enantiomers in plasma or brain would give rise to apparent time- and treatment-dependent shifts in the concentrationeffect relationship, when results based on assay of total enantiomer concentration are examined. Indeed, in our earlier efficacy study with the timed PTZ infusion rat seizure mode12*, a rightward shift in the ~ticon~ls~t effect-concentration curve was observed after subchronic administration when compared with acute administration of racemic stiripentol. Therefore, the apparent ‘tolerance’ phenomenon may not have been a true pharmacodynamic adaptation to chronic racemic stiri-

pent01 treatment, but may reflect a preferential a~cum~ation of the less active enantiomer in plasma and brain. If pharmacokinetic interactions between enantiomers were to occur, the disposition kinetics of each enantiomer following racemic stiripentol administration would have to be defined in order to appreciate fully the pharmacodynamics of the racemate. Indeed, our observation of a slower metabolic clearance of racemic stiripentol, when compared with either of its enantiome~, strongly suggests that stiripentol en~tiome~ can inhibit each other’s metabolism. Mutual inhibition of metabolism is plausible, since earlier studies have shown that cytochrome P450-mediated oxidative cleavage of the methylenedioxy ring leads to the formation of an inhibitory ligand complex between a reactive metabolite of stiripentol and the prosthetic heme moiety of the cytochrome enzyme(s)23. The situation may be much more complex than is realized in view of a recent abstract from Zhang et al.24 reporting the findings of metabolic conversion of (+)-stiripentol to its antipode in rats. In conclusion, there is stereoselectivity but not stereospecificity in the anticonvulsant action of stiripentol. The combined effect of the enantiomers appears to follow an additive interaction. Substantial differences in clearance kinetics were observed between the enantiomers. Moreover, the present data revealed the presence of a pronounced metabolic interaction between enantiomers after racemic drug administration. A detailed investigation of the differential pharmacokinetics of stiripentol enantiomers should be undertaken to further elucidate the role of stereoisomers in the pharmacodynamics of racemic stiripentol.

This work was supported by a grant from Laboratoires Biocodex. We would like to thank M. Moore for expert technical assistance.

36 References 1 Astoin, J., Marivain, A., Riveron, A., Crucifix, M., Lapotre, M. and Torrens, Y., Action de nouveaux alcools aCthyltniques sur le systtme nerveux central, Eur. 1. Med. Chem. Chim. Ther., 13 (1978) 41-47. 2 Babin, P., Lapouyade, P. and Dunogues, J., Synthesis of chalcone ethynologues with a pharmaceutical objective, Can. J. Chem., 60 (1982) 379-382. 3 Danzin, C., Claverie, N. and Jung, M.J., Stereochemistry of the inactivation of 4-aminobutyrate:2-oxoglutarate aminotransferase and t-glutamate I-carboxylase by 4aminohex-S-enoic acid enantiomers, B&hem. Pharmacol., 33 (1984) 1741-1746. 4 Gibaldi, M. and Perrier, D., Phormacokinetics, 2nd edn. revised, Marcel Dekker, New York, NY, 1982,494 pp. 5 Haegele, K.D. and Schechter, P.J., Kinetics of the enantiomers of vigabatrin after oral administration of the racemate or the pharmacologically active S-enantiomer, Clin. Pharmacol. Thu., 40 (1986) 581-586. 6 Haley, T.J. and Gidley, J.T., Pharmacological comparison of R(+), S(-) and racemic secobarbital in mice, Eur. J. Pharmacol., 9 (1970) 358-361. 7 Hildebrand, F.B., Introduction to Numerical Analysis, McGraw-Hill, New York, NY, 1956, pp. 80-85. 8 Holford, N.H.G. and Sheiner, L.B., Understanding the dose-effect relationship: clinical application of pharmacokinetic-pharmacodynamic models, Clin. Pharmacokinet., 6 (1981) 429-453. 9 Kohn, H., Conley, J.D. and Leander, J.D., Marked stereo-

10

11

12

13

specificity in a new class of anticonvulsants, Brain Rex, 457 (1988) 371-375. Kiipfer, A. and Branch, R.A., Stereoselective mephobarbital hydroxylation cosegregates with mephenytoin hydroxylation, C/in. Pharmacol. Ther., 38 (1985) 414-417. Kupferberg, H.J., Other hydantoins: mephenytoin and ethotoin. In: D.M. Woodbury, J.K. Penry and C.E. Pippenger (Eds.), Antiepileptic Drugs, Raven Press, New York, NY, 1982, pp. 283-295. Leander, J.D., Robertson, D.W., Clark, CR., Lawson, R.R. and Rathbun, R.C., Pharmacological effects of enanof 4-amino-N-(a-methylbenzyJ)benzamide, a tiomers chemically novel anticonvulsant, Epilepsia, 29 (1988) 83-90. Lin, H.-S. and Levy, R.H., Pharmacokinetic profile of a

new anticonvulsant, stiripentol, in the rhesus monkey, Epilepsia, 24 (1983) 692-702. 14 Loiseau, P. and Duche, B., Potential antiepileptic drugs stiripentol. In: R. Levy, R. Mattson, B. Meldrum, J.K. Penry and F.E. Dreifuss (Eds.), Antiepileptic Drugs, 3rd edn., Raven Press, New York, NY, 1989, pp. 955-%9. 15 Meldrum, B.S. and Murugaiah, K.M., Anticonvulsant action in mice with sound-induced seizure of the optical isomers of y-vinyl GABA, Eur. J. Pharmacol., 89 (1983) 149-152. 16 Moreland, T.A., Artoin, J., Lepage, F., Tombret, F., Levy, R.H. and Baillie, T.A., The metabolic fate of stiripentol in man, Drug Metab. Disp., 14 (1986) 654-662. 17 Orloff, M.J., Williams, H.L. and Pfeiffer C.C., Timed intravenous infusion of metrazol and strychnine for testing anticonvulsant drugs, Proc. Sot. Exp. Biol. Med., 70 (1949) 254-257. 18 Poisson, M., Huguet, F., Savattier, A., Bakri-Logeais,

F. and Narcisse, G., A new type of anticonvulsant, stiripentol: pharmacological profile and neurochemical study, Arz-

neim. Forsch., 34 (1984) W-204. 19 Pollack, G.M. and Shen, D.D., A timed intravenous

pentylenetetrazol infusion seizure model for quantitating the anticonvulsant effect of valproic acid in the rat, J. Pharma-

col. Methodr, 13 (1985) 135- 146. 20 Shen, D.D., Levy, R.H., Moor, M. and Savitch, J.L., Effi-

cacy of stiripentol in the intravenous pentylenetetrazol infusion seizure model in the rat, Epilepsy Res., J (1990) 40-48. 21 Tanno, N. and Terashima, S., Asymmetric reduction of various types of ketones with lithium aluminum hydride partially decomposed with (-)-N-methylephedrine and Nethylaniline, Chem. Pharm. Bull., 31(1983) 837-851. 22 Yeh, K.C. and Kwan, K.C., A comparison of numerical integrating algorithms by trapezoidal, lagrange, and spline approximation, J. Pharmacokinet. Biopharm., 6 (1978) 7998. 23 Zhang, K., Lepage, F., Cuvier, G., Astoin, J., Rashed, M.S. and Baillie, T.A., The metabolic fate of stiripentol in the rat: studies on cytochrome P-450-mediated methylene dioxy ring cleavage and side chain isomerism, Drug Metab. Disp., 18 (1990) 794-803. 24 Zhang, K., Lepage, F., Rashed, M. and Baillie, T.A., Stereoselective disposition of stiripentol in the rat, Phamtacolo@, 33 (1991) 234.